Don’t believe scientists who say all drinking is bad. Light boozing has strong benefits

Featured, News & Analysis

11th December 2015

Alcohol consumption has been a part of most human cultures since prehistoric times. While drinking practices vary by culture, ranging from total abstention to avid immersion, most adult drinkers are social drinkers who enjoy alcohol in moderation.

However, about 10-15 per cent of drinkers abuse alcohol. The most severe health effects of chronic heavy alcohol consumption include damage to the liver (alcoholic hepatitis or liver cirrhosis), heart (cardiomyopathy, hypertension), brain (structural changes, decreased cognition), peripheral neuropathy, and cancer, particularly in the digestive tract. Almost as old as the history of alcohol consumption are the efforts of governments, NGOs and, more recently, international agencies such as WHO, to reduce harmful alcohol consumption.

The picture is less clear regarding moderate alcohol consumption. On the one hand, countless studies have shown that moderate drinking is associated with health benefits, including decreased risk of coronary heart disease (heart attacks), ischemic stroke, type-2 diabetes, rheumatoid arthritis, improved cognition in the elderly and even a decreased risk of thyroid and kidney cancers (renal cell carcinoma) and non-Hodgkin lymphoma, resulting in decreased all-cause mortality. There is some discussion on what constitutes moderate drinking (generally assumed not to exceed an average of a few standard size drinks a day).

Needless to say, these potential benefits do not apply to everyone, and some people are at higher risk for damaging effects of alcohol consumption. These include, for instance, individuals infected with hepatitis B or C, or people who have genetic defects in the capacity to metabolise acetaldehyde, the primary product of alcohol metabolism (a defect found in 20 to 40 per cent of the population in East Asia). Also, there is general agreement that pregnant women should abstain from consuming alcohol to avoid the risk of damage to the developing foetus.

On the other hand, recent reports have highlighted the potential cancer risks of drinking even moderate amounts of alcohol; some scientists advocate that there is no safe level of drinking. For instance, several studies of the impact of alcohol consumption on breast cancer risk in women reported that an increase in risk can be precipitated by as little as a third of a drink a day. Similar increases in cancer risk could apply to other types of cancer. These reports have caused anxiety and led to recommendations that all alcohol use should be discouraged.

While caution is generally a good thing in the context of cancer risks, this needs to be balanced by the potential benefits of moderate drinking.

There are reasons to interpret the epidemiological studies that suggest a causative role of moderate alcohol consumption in breast cancer with considerable caution. By necessity, all of these studies rely on self-reporting to determine the amount of alcohol consumed. Although methods to assess self-reports with regard to recent drinking have become more standardised, they remain fraught with problems. Several recent studies demonstrate recall bias, where individuals substantially underestimate not only their ‘average’ daily consumption, but also the occasions when they exceed this average drinking pattern. Surprisingly, the degree of underestimation tends to be greater for moderate drinkers.

This problem is exacerbated when attempting to identify drinking patterns over extended periods. Breast cancer develops over decades; thus, correlations between alcohol consumption and breast cancer cannot be determined in studies with windows of alcohol exposure that capture only current or recent alcohol intake, after clinical diagnosis.

The apparent increased risk has been suggested to be substantially due to underreporting of intake. The purported precision of some of the outcomes of epidemiological studies that identify risks of less than one drink a day is to be treated with several grains of salt. Studies are needed that overcome this problem, with a prospective study design that follows individuals over prolonged periods with continued assessment of alcohol use, including estimates of underreporting.

A second problem is the lack of understanding of the mechanisms by which alcohol use could increase the risk for cancer. It is generally assumed that acetaldehyde, the primary alcohol metabolite, is the culprit and acetaldehyde is classified as a carcinogen by WHO. Although most individuals rapidly break down acetaldehyde formed from alcohol and maintain very low levels in blood, its local accumulation in tissues can be considered a risk factor. Acetaldehyde is also generated in other conditions. Smoking, for instance, increases the level of acetaldehyde in the mouth and throat and in the blood. This may be contributing to the finding of a higher than multiplicative risk for head and neck cancer in people who smoke and drink, often the majority of heavy drinkers. (That is, the risk for these individuals is much greater than if you simply combined the risks for smoking and drinking.)

smoking cigars and drinking whisky

An interesting ‘experiment of nature’ gives some insight into potential risks of acetaldehyde formation during alcohol consumption. Those people of East Asian origin who are genetically deficient in acetaldehyde metabolism accumulate much higher levels of this harmful compound in their blood when they drink. Usually, individuals with this genetic defect avoid alcohol consumption due to its unpleasant physiological effects; however, some of these still use alcohol, even to excess, often due to social pressure.

These individuals are at significantly increased risk for developing cancer of the oesophagus, but not for breast cancer. This suggests that if moderate alcohol use increases breast cancer risk, the mechanism may be quite different and identifying it would allow a better assessment of the nature of such risks.

Cancer is a dreaded disease and efforts to decrease its burden globally are strongly to be encouraged. However, the current state of knowledge of the association between alcohol use and breast cancer is ambiguous and confusing to both a woman and her doctor. High-quality, unbiased research into this important area is urgently needed, so that policies to reduce alcohol-induced harm can be based on sound science.

Women with known genetic susceptibilities for breast cancer should consult with their doctor about risk factors and are well advised to avoid overconsumption of alcohol. However, to stop moderate drinking to avoid the risks for cancer may do more harm than good.

Dr Jan B Hoek is a professor in the Department of Pathology, Anatomy and Cell Biology and Vice-Chair for Research at Thomas Jefferson University


  • scaz

    The missing million units of alcohol that duty is paid and nobody admits to drinking ? makes a nonsense the claimed risks of low or moderate consumption